Downregulation of Matriptase Inhibits Porphyromonas gingivalis Lipopolysaccharide-Induced Matrix Metalloproteinase-1 and Proinflammatory Cytokines by Suppressing the TLR4/NF-κB Signaling Pathways in Human Gingival Fibroblasts

Matriptases are cell surface proteolytic enzymes belonging to the type II transmembrane serine protease family that mediate inflammatory skin disorders and cancer progression. Matriptases may affect the development of periodontitis via protease-activated receptor-2 activity. However, the cellular mechanism by which matriptases are involved in periodontitis is unknown. In this study, we examined the antiperiodontitis effects of matriptase on Porphyromonas gingivalis-derived lipopolysaccharide (PG-LPS)-stimulated human gingival fibroblasts (HGFs). Matriptase small interfering RNA-transfected HGFs were treated with PG-LPS. The mRNA and protein levels of proinflammatory cytokines and matrix metalloproteinase 1 (MMP-1) were evaluated using the quantitative real-time polymerase chain reaction (qRT-PCR) and an enzyme-linked immunosorbent assay (ELISA), respectively. Western blot analyses were performed to measure the levels of Toll-like receptor 4 (TLR4)/interleukin-1 (IL-1) receptor-associated kinase (IRAK)/transforming growth factor β-activated kinase 1 (TAK1), p65, and p50 in PG-LPS-stimulated HGFs. Matriptase downregulation inhibited LPS-induced proinflammatory cytokine expression, including the expression of IL-6, IL-8, tumor necrosis factor-α (TNF-α), and IL-Iβ. Moreover, matriptase downregulation inhibited PG-LPS-stimulated MMP-1 expression. Additionally, we confirmed that the mechanism underlying the effects of matriptase downregulation involves the suppression of PG-LPS-induced IRAK1/TAK1 and NF-κB. These results suggest that downregulation of matriptase PG-LPS-induced MMP-1 and proinflammatory cytokine expression via TLR4-mediated IRAK1/TAK1 and NF-κB signaling pathways in HGFs.


Introduction
Periodontal disease is a serious inflammatory disease affecting the gingiva that can lead to tooth damage and other health complications. Periodontal disease is mainly caused by infection with Porphyromonas gingivalis (P. gingivalis) and is the primary cause of tooth loss owing to damaged soft tissues and bone resorption around the teeth [1,2]. P. gingivalis produces a lipopolysaccharide (PG-LPS) that represents the major virulence factor [3]. LPS produces various proinflammatory cytokines and mediators of inflammation such as interleukin-(IL-) 6, IL-8, tumor necrosis factor-α (TNF-α), IL-1β, prostaglandins, nitric oxide, and matrix metalloproteinases (MMPs) [4,5]. The underlying mechanism of the inflammatory response induced by PG-LPS relies on its interaction with Toll-like receptor-4 (TLR4). Additionally, PG-LPS upregulates MMPs and cytokines via various signaling pathways by binding to TLR4 [6][7][8]. Therefore, inhibiting signaling pathways associated with MMPs and cytokine expression may suppress periodontal disease [9].

Western Blotting Analysis.
HGFs were lysed using icecold radioimmunoprecipitation assay buffer (Thermo Scientific, Rockford, IL, USA). Cellular proteins (20 μg) were analyzed by sodium dodecyl sulfate-polyacrylamide gel electrophoresis and transferred to Hybond™ polyvinylidene fluoride membranes (GE Healthcare Life Sciences, Buckinghamshire, United Kingdom). Each membrane was blocked with a blocking buffer (5% bovine serum albumin or 5% skim milk) for 2 h at 4°C and then incubated with primary antibodies (1 : 2500 dilution) overnight at 4°C.
Antibodies against IL-1 receptor-associated kinase (IRAK1), TAK1, IKKα, and IKKβ, and the phosphorylated forms of the inhibitory subunit of NF-κBα (IκBα), IκB kinase αβ (IKKαβ), IRAK1, and TAK1 were purchased from Cell Signaling Technology (Beverly, MA, USA). Antibodies specific for proliferating cell nuclear antigen and horseradish peroxidase-conjugated secondary IgGs were obtained from Santa Cruz Biotechnology (Santa Cruz, CA, USA) and Cell Signaling Technology, respectively. Antibodies specific for matriptase were obtained from R&D Systems (Minneapolis, MN, USA). Antibodies specific for p50 and p65 were obtained from Abcam (Cambridge, UK). The blots were washed in Tris-buffered saline with 0.2% Tween-20 and then incubated with secondary horseradish peroxidaseconjugated goat anti-mouse or anti-rabbit for 1 h at 4°C. The β-actin antibody was from Sigma-Aldrich. β-Actin and proliferating cell nuclear antigen were used as loading controls. Protein expression levels were determined using a Mini HD6 Image Analyzer and Alliance 1D (UVItec Cambridge; Cleaver Scientific, Rugby, United Kingdom).  2.6. Enzyme-Linked Immunosorbent Assay (ELISA). MMP-1 in conditioned medium was quantified using a sandwich ELISA kit (cat. #F1M00, R&D Systems) according to the manufacturer's protocol. HGFs were transfected with matriptase siRNA and stimulated with PG-LPS for 24 and 48 h at 37°C. Standards and sample were added and incubated at room temperature for 3 h. After 4 washes, APMA (p-aminophenylmercuric acetate) was added to the stan-dards and sample wells for 2 h at room temperature. After 4 washes, substrate was added to each well for 17~20 h at 37°C. The reaction was determined (RFU) using a microplate reader (Sunrise™, Tecan Group, Männedorf, Switzerland). MMP-1 activation was given as the active MMP-1 levels (ng/mL) in the cell culture medium. Absorbance at 450 nm was measured using a microplate reader (Sunrise™, Tecan Group, Männedorf, Switzerland).

Analysis of Nuclear and Cytoplasmic Extracts.
HGFs were transfected with 100 pmol siRNA against matriptase or negative control siRNA using the RNAiMAX Transfection Reagent (Thermo Fisher Scientific) for 24 h, then incubated with PG-LPS for 3 h. Nuclear and cytoplasmic proteins were extracted using the NE-PER Nuclear and Cytoplasmic Extraction Kit (Pierce, Rockford, IL, USA), according to the manufacturer's instructions. The protein concentrations in the membrane, cytosol, and nuclear fractions were determined using a protein assay.

Statistical
Analysis. Data are expressed as means ± standard error. Statistical significance was determined using a one-way analysis of variance followed by the Scheffe post hoc test using Microsoft Excel (Redmond, WA, USA). A value of p < 0:05 and p < 0:01 represent significant differences. All experiments were performed in triplicate.

Effect of Matriptase Downregulation on PG-LPS-Induced MMP-1 Expression in HGFs.
To investigate the effect of matriptase on PG-LPS-induced MMP-1 expression, we first confirmed that matriptase expression was downregulated by approximately 2.7-fold compared to control siRNA after transfection with matriptase siRNA (Figure 1(a)). The MMP1 mRNA expression in HGFs was analyzed by the  (Figure 1(b)). The ELISA confirmed that PG-LPS treatment of HGFs resulted in an increase in MMP-1 protein secretion, while matriptase siRNA significantly diminished this increase by approximately 1.7-fold at 48 h compared to control siRNA (Figure 1(c)). These results indicate that matriptase involved in PG-LPS-mediated MMP-1 expression.

Effect of Matriptase Downregulation on the Expression of PG-LPS-Induced Proinflammatory Cytokines in HGFs.
Previous studies have suggested a pivotal role of proinflammatory cytokines in periodontal diseases [25,36]. Therefore, we evaluated the effects of matriptase downregulation on proinflammatory cytokine expression in HGFs. Matriptase siRNA-transfected HGFs were exposed to PG-LPS for 6 h and then the mRNA was extracted and analyzed. PG-LPS stimulation dramatically increased mRNA expression levels by approximately 11.4-fold for IL6, 116.1-fold for IL8, 31.5-fold for TNFα, and 2679.1-fold for IL1β compared to control siRNA, whereas matriptase siRNA-transfected HGFs showed considerably decreased by approximately 9.2-fold for IL6, 80.7-fold for IL8, 6.7-fold for TNFα, and 242.8-fold for IL1β (Figure 2). These results indicate that matriptase involved in PG-LPS-induced proinflammatory cytokine expression.

Downregulation of Matriptase
Inhibited PG-LPS-Induced IRAK1/TAK1/NF-κB Activation in HGFs. The binding of LPS to TLR4 increases MMP and proinflammatory cytokine expression via various signaling pathways such as IRAK, TAK, and NF-κB during the development of periodontitis [13,16,37]. Therefore, to investigate the effects of matriptase downregulation on LSP-PG-induced IRAK1, TAK1, and NF-κB activation, we transfected HGFs with matriptase siRNA. Matriptase siRNA reduced the phosphorylation of IRAK1 and TAK1 at 5 min after PG-LPS treatment (Figure 3(a)). Additionally, matriptase siRNA transfection suppressed the phosphorylation of IKKαβ and IκBα in the cytoplasmic fraction and reduced the nuclear translocation of NF-κB p50 (Figures 3(b) and 3(c)). These results indicate that matriptase was involved in PG-LPSinduced TLR4 signaling-mediated MMP-1 and proinflammatory cytokine expression in HGFs.  BioMed Research International Figure 1). Thus, these results indicate that matriptase was involved in PG-LPS-mediated TLR4 expression. Next, to elucidate the role of the TLR4 signaling pathway in the response of HGFs to PG-LPS, we treated these cells with M62812, a TLR4 inhibitor. We then determined the phosphorylation or expression of certain key components of the TLR4 pathway including IRAK, TAK1, IKK, IκBα, and NF-κB (subunits p65 and p50). M62812 treatment reduced p-IRAK, p-TAK1, p-IKKαβ, and p-IκBα levels in the cytoplasmic fraction and nuclear translocation of NF-κB p50 (Figures 4(c) (Figures 4(a) and 4(b)). These findings suggest that PG-LPS stimulation regulates MMP-1 and proinflammatory cytokine expression via TRL4-mediated IRAK1/TAK1 and NF-κB activation in HGFs.  (Figures 5(a) and 5(b)). The treatment additionally attenuated PG-LPS-mediated IRAK1/TAK1 signaling ( Figure 5(c)). These results indicate that PAR-2 activity was involved in PG-LPS-induced MMP-1 and proinflammatory cytokine expression in HGFs.

Discussion
Previous studies suggested that matriptase plays an important role in the pathogenesis of periodontal disease caused by P. gingivalis. However, the cellular mechanisms via which matriptase mediates periodontal disease have not been studied in detail. Our results demonstrated the role of matriptase on PG-LPS-induced MMP-1 and inflammatory cytokine  Periodontal disease is the most common inflammatory disease and one of the main causes is P. gingivalis infection [38,39]. LPS, a major component of the outer membrane in P. gingivalis, binds to one of the TLRs, TLR4, which is one of the main virulence factors that induce inflammation [18,40]. The stimulation of HGFs by PG-LPS activates multiple TRL4-mediated signaling pathways [15,18,41]. The binding of LPS to TLR4 results in the recruitment of MyD88, which forms a complex with IRAK and another TIR-containing adapter molecule, Mal (similar to the MyD88 adapter), followed by the activation of TAK1 [15][16][17]42]. This activation leads to activation of the IKK/ NF-κB pathways [42,43]. In particular, IKK activation induces the expression of MMP and proinflammatory cytokines via phosphorylation of IκBα and translocation of NF-κB subunits to the nucleus [18,44]. These proinflammatory cytokines and MMP can induce periodontal tissue injury [21,22,45]. To clarify the signaling mechanism of TLR4, we demonstrated the inhibitory effect of a TLR4 inhibitor (M62812) on PG-LPS-induced MMP-1 and proinflammatory cytokine expression in HGFs (Figure 4).

BioMed Research International
Previous studies have indicated that MMPs play major roles in the degradation of the bone matrix and ECM in periodontal disease development [46,47]. In active periodontal disease, the progressive destruction of periodontal tissue and alveolar bone is increased through the expression of MMPs and inflammatory cytokines [45,47]. Various MMPs, such as MMPs-1, 2, 3, 8, 9, and 13, are involved in the loss of periodontal tissue [21,22,48]. Among these MMPs, MMP-1 is the main component of the periodontal tissue matrix and plays a major role in periodontal ECM degradation and remodeling [21,48,49].
In periodontal diseases, proinflammatory cytokine activity is regulated by bacteria and their products [50][51][52]. IL-6 plays an important role in continuous tissue destruction and bone resorption via osteoclast differentiation in the infected periodontal site [52]. IL-8 induces neutrophil migration to periodontal lesions. This can weaken periodontal tissues by generating intracellular reactive oxygen species and increasing MMP expression and the release of lysosomal enzymes [53,54].
During periodontitis pathogenesis, IL-1 is involved in the inflammatory response and ECM remodeling through increased expression of various factors including cytokines, MMPs, reactive oxygen species, nitric oxide synthase, and prostaglandins [55][56][57]. Moreover, it is well-known that stimulation of HGFs with IL-1β produces IL-6, IL-8, and TNF-α [58,59]. TNF-α is secreted primarily by immune cells or fibroblasts, and it induces the production of MMPs, cytokines, prostaglandin E2, cell adhesion molecules, and factors involved in bone resorption [60,61]. Therefore, because the regulation of MMP-1 and proinflammatory cytokines is important for the treatment of periodontitis, we demonstrated that downregulating matriptase inhibited the PG-LPS-induced MMP-1 and proinflammatory cytokine expression in HGFs (Figures 1 and 2).
Currently, existing matriptase inhibitors include monoclonal antibodies, peptide-based inhibitors, and small molecule inhibitors, and these have been mainly developed as anticancer drugs. However, the current development of chemical and biochemical matriptase inhibitors is insufficient. The aim of this study was to improve the development of agents that suppress periodontal disease through an enhanced understanding of how to regulate various critical mechanisms. Matriptase regulates multiple intracellular signaling pathways by cleaving the activation site of PAR-2, a G protein-coupled receptor [28][29][30]. In previous studies, PAR-2 was implicated in the pathogenesis of periodontal disease caused by P. gingivalis [33,34]. Furthermore, it has been reported that signaling via the interaction of PAR-2 and TLR4 induces MMP and cytokine expression [70][71][72]. Nevertheless, the cellular mechanism via which matriptase regulates the MMPs and proinflammatory cytokines involved in periodontitis was unknown. Therefore, we confirmed the effect of a PAR-2 inhibitor (GB83) on the signaling pathway of PG-LPS-induced MMP-1 and proinflammatory cytokine expression in HGFs ( Figure 5). Additionally, we confirmed the inhibitory effect on the signaling pathway of PG-LPSinduced MMP-1 and proinflammatory cytokine expression via the downregulation of matriptase in HGFs (Figure 3).

Conclusions
Our study demonstrated the signal transduction mechanism by which matriptase affects PG-LPS-induced expression of MMP-1 and proinflammatory cytokines in HGFs. Our data showed that decreasing matriptase inhibited PG-LPSinduced MMP-1 and proinflammatory cytokine expression via TLR4-mediated IRAK1/TAK1 and NF-κB signaling pathways in HGFs. These results provide insights into the therapeutic potential of matriptase for preventing and treating periodontal disease.

Data Availability
The data used to support the findings of this study are included within the article.

Conflicts of Interest
The authors declare that they have no conflict of interests.